Bored this weekend? Consider joining me in sprinting to make something impressive with open models for our competition, 4 winners get a giant 4.5kg Toblerone chocolate bar

https://redd.it/1ola2p2
@rStableDiffusion

Qwen3-VL support merged into llama.cpp
https://github.com/ggml-org/llama.cpp/pull/16780

https://redd.it/1olenpo
@rStableDiffusion

My latest consistency test

https://redd.it/1olgq14
@rStableDiffusion

Introducing InScene + InScene Annotate - for steering around inside scenes with precision using QwenEdit. Both beta but very powerful. More + training data soon.

https://redd.it/1olgsxr
@rStableDiffusion

Ollama's engine now supports all the Qwen 3 VL models locally.

Ollama's engine (v0.12.7) now supports all Qwen3-VL models locally! This lets you run Alibaba's powerful vision-language models, from 2B to 235B parameters, right on your own machine.

https://preview.redd.it/zwbgca6k6lyf1.png?width=4020&format=png&auto=webp&s=c38fb52d76b672a31462cb756b13841d0d2f7351



https://redd.it/1olgkf1
@rStableDiffusion

Created a free frame extractor tool

I created this Video Frame extractor tool. It's completely free and meant to extract HD frames from any videos. Just want to help out the community, so let me know how i can improve this. Thanks

https://redd.it/1olc7k5
@rStableDiffusion

Question about Training a Wan 2.2 Lora
https://redd.it/1ollb1b
@rStableDiffusion

Wow! The spark preview for Chroma (fine tune that released yesterday) is actually pretty good!

https://redd.it/1olm6ng
@rStableDiffusion

Update to my Synthetic Face Dataset

https://redd.it/1olmvxg
@rStableDiffusion